Knowledge Resource Center for Ecological Environment in Arid Area
| 大气二氧化碳反演与误差校正方法研究 | |
| 其他题名 | Retrieval Method and Error Correction of Atmospheric Carbon Dioxide |
| 叶函函 | |
| 出版年 | 2013 |
| 学位类型 | 博士 |
| 导师 | 方勇华 |
| 学位授予单位 | 中国科学院大学 |
| 中文摘要 | 二氧化碳（CO2）是大气中最重要的温室气体，对全球气候变化产生着重要的影响，因此准确掌握大气中CO2的含量及其变化就成为研究气候变化的基础。大气CO2地面观测虽然精度高，但观测站点分布稀疏难以满足源与汇研究的需要，而卫星观测以全球覆盖率和高频率采样的特点，可以实现全球观测与数据快速更新，因此近年来基于卫星平台的大气CO2观测技术得到迅速发展，目前，国内外已有多颗专用于大气CO2及其它温室气体探测的卫星在轨运行或将要发射。为满足气候研究的需求，卫星探测CO2浓度月平均结果的精度在1000*1000 km^2区域尺寸上要优于1%，这远高于当前水平，因此实现卫星数据的高精度反演，成为目前该领域的一项重要研究课题。 根据大气CO2的光谱变化特征，选择温室气体观测卫星（GOSAT）近红外波段光谱作为数据源。在系统构建的大气CO2反演方法的基础上，实现优化估计方法控制误差传播的模拟研究、反演过程中误差参数影响的校正，以及利用卫星实测数据进行的反演及误差校正方法评估。 基于贝叶斯定律的优化估计方法是处理病态大气遥感问题行之有效的方法，相比于确定性求解法，它可以通过设置合理的先验约束，有效地控制观测误差对反演结果的影响。然而由于先验约束难以准确构建，因此根据已有经验，提出简易构建方法，并利用带噪声的观测数据进行模拟分析，结果表明该构建方法合理且可有效地控制误差传播。 地表反射率和大气状态参数是卫星数据反演中决定结果精度的重要因素，为降低反演结果对其依赖程度并提高反演精度，分别提出比值光谱法和O2 A吸收带校正法。超光谱的反演中，off-line通道的选择是比值光谱法的关键，正确的off-line通道可使得近似线性误差源的校正效果最好，且不易受太阳光谱、水汽及随机噪声的影响。利用O2 A吸收带的信息将CO2柱含量转换成混合比，可较好地抑制大气状态参数、地表高度及大气分层方法带来的系统误差，而这种系统误差单独利用CO2吸收带是难以克服的。 利用2009年7月至2010年12月的塔克拉玛干沙漠GOSAT实测数据进行反演和误差校正，与GOSAT L2级产品及CarbonTracker地面同化数据进行比较，结果表明比值光谱法抑制了地表反射率的影响，使得反演结果的离散性大大减小，同时O2 A吸收带校正法也很好地控制了大气状态参数的影响，在此基础上，合理的先验约束进一步抑制了其它已知系统误差的影响，最终得到的月平均CO2混合比结果与实际较一致，且具有相似的年增长趋势及季节变化特征。对实际状态把握越准确，误差控制越明显，反之效果略差，显示出了大气CO2反演精度对反演方法、校正技术以及必要的先验知识的依赖。 |
| 英文摘要 | Since Carbon Dioxide (CO2) is the most important greenhouse gas in the atmosphere and plays an important role in global climate, accurate knowledge about atmospheric concentration and variability of CO2 is a foundation for study on climate change. Ground-based measurements of atmospheric CO2 are highly accurate, but have limited spatial coverage and hard to estimate the source and sink of CO2, while satellite-based observation can monitor the global distributions of greenhouse gases with high density and high spatiotemporal resolution, which motivates recent rapid development of satellite-based remote sensing techniques. Nowadays，many satellites dedicated to atmospheric CO2 and other greenhouse gases are already on orbit or may be launched in the further. In order to study the climate change, a precision of 1% for monthly mean CO2 column abundance on a regional scale (~1000*1000 km2) is required, which is substantially higher than existing situation. Therefore, achieving such high precision becomes a major goal in this area. Near-infrared spectra obtained by the Greenhouse gases Observing SATllite (GOSAT) was chosen as data source due to CO2 spectroscopic consideration. On the basis of systematic construction of atmospheric CO2 retrieval method, simulation study about the error controlling of optimal estimation method, correction of error caused by imperfect parameters and evaluation of retrieval method and error correction with satellite measurements were performed. Compared with least squares method, optimal estimation method based on Bayes principle is more suitable for ill-conditioned atmospheric retrieval problem, which has potential for controlling the retrieval error through a priori constraint. However, it is hard to construct theoretical constraints. For this reason, a simple and easy method based on existing experience was proposed, and simulation using observation with noise was performed. The simulated results demonstrate that error can be controlled effectively by this method. Surface albedo and atmospheric parameters are two important factors in satellite data retrieval. In order to reduce the error caused by imperfect parameters and improve the retrieval precision, ratio spectrometry method and O2 A band correction method were proposed accordingly. For high-spectral-resolution measurements, error correction ability of spectrometry ratio method depends significantly on the choice of off-line channel. Reasonable off-line channel can correct the linear error and not sensitive to random noise in observation and uncertainties in solar spectrum and atmospheric H2O. For errors caused by atmospheric parameters, surface height and atmospheric layers, it is hard to overcome by CO2 band separately, however, it has similar effect on Oxygen (O2) measurements, so it can be mostly corrected by employing O2 column abundance to transfer CO2 column abundance to volume mixing ratio. GOSAT spectra measured over TKLMG desert from June 2009 to December 2010 were used to test and validate the retrieval method and error correction. Compared to GOSAT L2 products and CarbonTracker ground-based assimilation data, the results show that the retrieval error caused by surface albedo and atmospheric parameters are corrected largely by ratio spectrometry method and O2 A band correction method separately, based on this, other known systematic error are controlled further by proper priori constraint, which makes monthly averaged CO2 mixing ratio nearly consistent with the references, and also has similar annual increase and seasonal variability. The error analysis has shown that more accurate knowledge about the reality, more significant improvements could be achieved. Nevertheless, the correction will be less effective. This validation also demonstrates the importance of retrieval method, correction technique and necessary priori constraints on CO2 retrieval precision. |
| 中文关键词 | 大气CO2 ; 反演方法 ; 误差校正 ; 验证 |
| 英文关键词 | atmospheric carbon dioxide retrieval method error correction validation |
| 语种 | 中文 |
| 国家 | 中国 |
| 来源学科分类 | 光学 |
| 来源机构 | 中国科学院合肥物质科学研究院 |
| 资源类型 | 学位论文 |
| 条目标识符 | http://119.78.100.177/qdio/handle/2XILL650/287252 |
| 推荐引用方式 GB/T 7714 | 叶函函. 大气二氧化碳反演与误差校正方法研究[D]. 中国科学院大学,2013. |
| 条目包含的文件 | 条目无相关文件。 | |||||
| 个性服务 |
| 推荐该条目 |
| 保存到收藏夹 |
| 导出为Endnote文件 |
| 谷歌学术 |
| 谷歌学术中相似的文章 |
| [叶函函]的文章 |
| 百度学术 |
| 百度学术中相似的文章 |
| [叶函函]的文章 |
| 必应学术 |
| 必应学术中相似的文章 |
| [叶函函]的文章 |
| 相关权益政策 |
| 暂无数据 |
| 收藏/分享 |
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。
